What is wrong with this LL ?

[itsmoked]

I have been struggling with this for 2 days now through various iterations. This logic is supposed to work this way: When the upper left contact closes the two timers become a blinker. Blinking the "LAMP". The counter counts the blinks and on the third one latches the LAMP on permanently.

I got the working one running just fine. (happened to be in the 2nd segment, 4th network)

I copied the entire network and pasted it in as the 4th network in the 1'st segment. I then changed all the logic to be 1XX instead of the original 2XX with the exception of the physical LAMP coils which are PLC outputs.

The 1XX network NEVER blinks and certainly never latches on.

It's like the timers 40111 & 40112 just don't work at all..
Why? Any pointers?

Works:

Busted:

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[waross]

Hi Keith
I would start with another lamp output with a lamp attached.
Use the left hand bus as the input.
Should light. If it doesn't, investigate why this network is inactive.
Next go to the right hand side of 00105 and prove circuit action there.
You basically have a stick that is compatible with your PLC, and you can poke your circuit until you find out where you are losing your signals.
(What is the action of your timers?)
Respectfully

 

[itsmoked]

[green]I would start with another lamp output with a lamp attached.
Use the left hand bus as the input.
Should light. If it doesn't, investigate why this network is inactive.[/green]
I did this already and the output works fine.

[green]Next go to the right hand side of 00105 and prove circuit action there.[/green]
Here is an online snapshot that shows the left timer should be doing a count.. it just sits there.

[green]You basically have a stick that is compatible with your PLC, and you can poke your circuit until you find out where you are losing your signals.[/green]
I have been poking, beating and bludgeoning to no avail. This is at a point where in other specialties I would be saying "get out of the box" cuz it doesn't feel right.

[green](What is the action of your timers?)[/green]
Non-functional! lolo

The left timer is supposed to count up to 0.3 sec then turn on its output. This enables the right timer which counts up to 0.3 sec. When it does it opens 113 which resets the first timer which resets the second timer. The cycle repeats.

Here are the instructions out of the manual.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[ScottyUK]

Keith,

First thought: make an exact copy of the original ladder and see if that works with the new algorithm numbers. Make your changes slowly and methodically.

Why not establish a 'flasher circuit' in one ladder and use the output in multiple locations, gating it as appropriate, rather than building multiple flashers? This is how it's normally done. Simple enhancements are a slow flash and fast flash, or flash/antiflash in opposite phase.

Does the PLC have a flash or heartbeat as a bit within the system registers? I can't remember - this is a Modicon I guess?

----------------------------------

I don't suffer from insanity. I enjoy it...

 

[waross]

Hi Keith
I understand your frustration.
I would be rechecking my addressing. Yup, I know, you've done that 56 times. I would have by now.
Have you checked for any ones transposed as zeros or vice versa?
Have you checked to see if your addresses are in use elsewhere in the segment?
It's a small circuit, how about entering it manually (or even one timer) instead of cut and paste?
Try renumbering the timers with the original numbers. 40211 and 40212. The timers may not have to be numbered as the segment.
How about redefining the inputs to the counter.
Connect 00105 to the count terminal where 00114 is presently connected and connect the enable input directly to the left hand line. See if it will count. You will have to reset it with the programer.
Your circuit worked in another segment, and it looks good. It must be an addressing issue. If you can make the counter count, it is a redundant check that the network is scanning.
Respectfully

 

[richs]

Hi-

Sorry I don't have time to really sit down and puzzle this
one out right now. At first blush, I would be looking
at what the enable and reset lines of the timers are doing!

If that reset toggles/glitches, you are hosed and it will never
complete the timer operation.

Again, this is first blush. I could be way out of line and
I apologize, OTOH, it MIGHT be the problem and you can
fix it right away.

Cheers,

Rich S.

 

[itsmoked]

Okay Gents;

Waross; I tried erasing the bad one and pasting the good one in. Then I carefully changed all addresses. No Joy!

Then I carefully checked to make sure I wasn't duplicating any addresses in other sections. No joy.

Then I changed the addresses to entirely different numbers. No Joy.

I replaced 113 with a short. Right timer never ran. No Joy.

Rich I don't see where a "glitch" comes in. Either this PLC is deterministic or I'm doomed. The outputs are all levels so "glitches" shouldn't even matter. Even if one was present it should clear up on the next scan.

I think something is broken....

Scotty your "one flasher circuit" idea is wonderful! I implemented it and boy did it whack out a truck load of logic! I ditched the "bad code" and even the "good code" and have things working well for the moment. Thanks!

I still, like probably most engineers, want to know why the 'obvious' didn't work! But for the moment I have a working solution.

Thanks for your help guys.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[djs]

Hello;
This maybe a stupid question but is segment scheduled to be executed? If you have segment 2 scheduled and not segment 1, then you will get the responses shown.

 

[itsmoked]

Hi djs;

I have no idea! I have never come across any mention of "scheduling" in my 'tool'. I have 13 sections and 15 networks scattered about in them and others work. Initially the first section worked fine, then I made the second section which didn't work right. Then I tore down the second one and remade it differently. It worked. I modified it and got all the features working.

Then I modded the first section to work the same as the 2nd. It wouldn't! So I cut and pasted several times the 2nd to the first. Still wouldn't work. Then I changed it all to something else (as per Scotty) and they both work... But I never disabled a section that I know of.. Very puzzling.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[waross]

Hi Keith;
I have probably misread you, but I'm wondering about cut and paste. To me that brings up images of a word processor type operation.
Is there a method to call up new devices rather than copying them? There may be some behind the scenes operations carried out by the routine that calls up new devices that don't happen when you cut and paste.
Respectfully
Bill

 

[itsmoked]

That's one of the advertised "features" of my tool: ProWORXs.

I take your point. I did 're-part'. To no avail.

I am now going to take my two good working channels and dupe them into the other eight positions. I'll report back.


Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[richs]

Hiya-

O.K. What I was trying to say that the input 105 of
Network 4/22 MUST remain high. If it goes low, it looks
like it will reset the timers again.

Timers are somewhat PLC type dependant with subtle
differences between vendors.

I've been looking at the AB type timers as referenced in
the book mentioned in the other thread that can be found at:

http://claymore.engineer.gvsu.edu/~jackh/books/plcs/

Great little book and the price is RIGHT!

Anyway, the AB type timers flasher example is on page
194 *OR* PLC Timers- 9.14.

As an aside, there are TON and TOF and TONR and TOFR types
with the R denoting Retentive timers (ones that don't
reset).

Cheers,

Rich S.

 

[itsmoked]

Hi Rich,

I shorted 105 many times without success. I changed the time periods up to a second to make sure that ONLINE mode could actually depict the cycling. hmmm.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Ashereng]

Keith,

I think what djs is saying is that on some systems, each "block" has a scheduleable setting, to set it to run in a particular sequence, or if left blan, whenever during available time on the execution cycle. It is a very common HWL thing dating back to its TDC days.

Or, I could be totally off my rocker.

djs, what say you?

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[waross]

Any luck yet, itsmoked? We haven't forgotten you.
We are all wondering if you solved the problem or just dropped the PLC down an elevator shaft.
respectfully

 

[itsmoked]

Hi waross;

No worries mate. I have changed to Scotty's method which essentially side stepped the issue by removing the errant evile code.

I still hate these <(mysteries)> though. I expect it has nothing to do with the logic but rather the tool. The answer will either never come or will be provided by someone who better understands the tool and responds with something like "oh yeah, you can't cut and past then re-label networks on even days of the month".

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[LionelHutz]

I know what you mean about mysteries. I was using a Zelio relay which is one of the small programmable ones where you enter ladder logic right on the display of the unit. I accidentally found that if I draw a connecting line over top of one that already exists it works fine till it's powered down then it "crashes" and clears the logic when it powers up again. It took a few experiments to figure that one out. Lots of fun.

 

[itsmoked]

Oh wow that is sooooo vile.. I run into that stuff with C compilers I just changed this one instruction and now the compiler doesn't compile six lines anymore that are 500 lines away argh!!

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[itsmoked]

Ah the saga continues.

I had the same sort of cascaded timers in several other logical networks. They blink the error lamp at different rates to multiplex the idiot light. Had the same exact problem.. First timer never starts timing up. Just sits there with enable set high(to allow counting). After considering different ways to smash the PLC into unidentifiable mush I swapped out the first timer which was a 0.1sec one initialized to 40 => 4.0 seconds I ripped it out and put in a 1.0sec timer and initialized it to 4 => 4 seconds. It worked fine.. no other changes..

Must be ProWORX screwing up ..?..?..!...!..?..&...&^%...%$@*%^..*%$^#@!)*&!!@#!!!

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com